voidly
KP · per-country backtest

North Korea forecast vs reality

Every Voidly Sentinel shutdown-risk forecast for North Korea, plotted against what actually happened. 73 (predicted, observed) pairs from the rolling 30-day evaluation window.

Updated every 30 min · CC BY 4.0 · Raw JSON · Current forecast →

Forecasts evaluated
73
since Apr 17
Accuracy @ 0.5
87.7%
64/73 correct
Brier score
0.002
lower is better
Observed positive rate
0%
mean predicted 4%

Forecast time series

Blue line: forecast probability. Green ✓ markers: forecast was right. Red ✗ markers: forecast was wrong. Dashed line at 0.5 is the binary decision threshold.

0.000.250.500.751.00threshold 0.50Apr 17Jun 28

Y axis: forecast probability · Faint tick: distance to observed outcome (0 or 1)

All 73 predictions (newest first)

Eval dateForecastPred ≥ 0.5?Observed?Correct?
Jun 284.3%—no event✓
Jun 274.1%—no event✓
Jun 263.3%—no event✓
Jun 254.4%—no event✓
Jun 244.8%—no event✓
Jun 234.9%—no event✓
Jun 224.8%—no event✓
Jun 213.5%—no event✓
Jun 203.7%—no event✓
Jun 195.3%↑no event✗
Jun 184.2%—no event✓
Jun 174.0%—no event✓
Jun 162.8%—no event✓
Jun 154.6%—no event✓
Jun 144.4%—no event✓
Jun 134.9%—no event✓
Jun 122.9%—no event✓
Jun 114.2%—no event✓
Jun 105.1%↑no event✗
Jun 93.0%—no event✓
Jun 84.4%—no event✓
Jun 74.6%—no event✓
Jun 63.2%—no event✓
Jun 54.7%—no event✓
Jun 44.1%—no event✓
Jun 34.4%—no event✓
Jun 24.2%—no event✓
Jun 13.9%—no event✓
May 313.7%—no event✓
May 304.5%—no event✓
May 293.4%—no event✓
May 282.7%—no event✓
May 274.1%—no event✓
May 263.8%—no event✓
May 255.5%↑no event✗
May 243.9%—no event✓
May 234.3%—no event✓
May 223.7%—no event✓
May 215.7%↑no event✗
May 205.5%↑no event✗
May 194.8%—no event✓
May 185.2%↑no event✗
May 172.3%—no event✓
May 165.0%↑no event✗
May 153.8%—no event✓
May 143.4%—no event✓
May 134.8%—no event✓
May 123.8%—no event✓
May 113.3%—no event✓
May 104.0%—no event✓
May 95.2%↑no event✗
May 83.4%—no event✓
May 73.9%—no event✓
May 63.9%—no event✓
May 54.4%—no event✓
May 43.3%—no event✓
May 32.6%—no event✓
May 24.8%—no event✓
May 14.8%—no event✓
Apr 304.6%—no event✓

How to read this

  • Each row is one historical forecast. We made the prediction at eval_date, then waited the 7-day horizon, then graded against the observed outcome.
  • Forecast % is the calibrated probability we published that day. Post-recalibration (2026-05-20) these now match actual observed rates much better — see the refit finding.
  • Pred ≥ 0.5 shows the binary alert decision. Note we usually fire alerts at a lower threshold (see /v1/sentinel/global_heatmap for the live cutoff) — the 0.5 column here is for backtest scoring consistency with the global confusion matrix.
  • Correct = both (pred ≥ 0.5) and (observed) agree.

Related